一种基于全变差正则化与小波包变换的图像去噪算法

摘要

Authors proposed an effcient image denoising method based on the combination of wavelet packet transform with total variation model and presented how to select the regularization parameter in this model.The combination of wavelet packet transform with total variation model helps to alleviate staircase effect efficiently and preserve sharp discontinuities in images as well.The numerical experimental results show that the new method is effective in removing Gaussian noise and keep the detail of the image well.%提出一种基于全变差(TV)模型和小波包变换的图像去噪算法，并给出了针对该模型的一种改进正则化参数选取方法，改善了全变差模型去噪中出现的块效应问题，同时保留了图像中的边缘信息。数值实验表明，用所给算法去噪可得到较高的峰值信噪比和较好的视觉效果。